Bertin Ergonomie facteur humain is pleased to announce the thesis defence of Mr Robin Eyraud (and invites you to attend).
Robin completed his thesis with our team over a period of 3 years, working directly on our Research & Development projects for the automotive sector (Advanced Driver Interaction field) and in partnership with the University of Paris 8, the academic partner for this research, under the supervision of Thierry Baccino and Elisabetta Zibetti.
Subject of thesis :
Allocation of visual attention while driving: the impact of augmented reality
Venue and time of presentation :
Cité des sciences
Friday 23 December at 1pm
Salle Jean Painlevé (library, level -2)
Summary of the thesis:
Attention is the process of filtering information that is useful to an activity from that which is not. Highlighting augmented reality (AR) guides this information selection process, highlighting certain elements over others. As currently envisaged for car driving, highlighting AR highlights elements linked to the general driving activity (e.g. signs in poor visibility, direction to take), but independently of manoeuvres. However, the literature on visual attention in activity shows us that eye pathways are highly specific to immediate goals and sub-goals. AR that does not respect this «natural» prioritisation of information processing is therefore likely to disrupt information acquisition.
The first objective of this research is to determine the extent to which the allocation of visual attention while driving focuses on information related to the manoeuvre. The second objective is to study the impact of AR on this allocation of attention.
We set up three experiments in which participants viewed static and dynamic driving scenes and had to decide whether they could perform a manoeuvre. We analysed variations in the allocation of visual attention according to the manoeuvres notified and the RA conditions using oculometric recordings. Our results show that visual attention is strongly allocated to cues for decision making, but does not neglect cues for general understanding of the scene. AR optimises visual attention when it highlights manoeuvre-related cues, but disrupts visual attention under other conditions. These results make it possible to identify and characterise the various risks inherent in highlighting AR, and to discuss design options for taking them into account.